1. Field of the Invention
The present invention relates to oligomerization of olefins to form higher olefins, especially higher xcex1-olefins. The instant invention also relates to the use of sulfur-containing and sulfur-tolerant catalysts to perform the oligomerization.
2. Description of the Related Art
Linear xcex1-olefins are very important and versatile intermediates and building blocks for the chemical industry. The main applications for linear xcex1-olefins are using C4-C8 xcex1-olefins as co-monomers for polyethylene, C6-C10 xcex1-olefins as feedstocks for plasticizers, and C12-C20 xcex1-olefins for surfactants. Hydrocarboxylation of the C6-C8 xcex1-olefins with cobalt carbonyl/pyridine catalysts gives predominantly linear carboxylic acids. The acids and their esters are used as additives for lubricants. The C6-C10 xcex1-olefins are hydroformylated to odd-numbered linear primary alcohols, which are converted to phthalate esters (by reaction with phthalic anhydride) as polyvinyl chloride (PVC) plasticizers. Oligomerization of 1-decene, using BF3 catalysts, gives oligomers used as synthetic lubricants known as poly-xcex1-olefins (PAO) or synthetic hydrocarbons. The C10-C12 xcex1-olefins can be epoxidized by peracids; this opens up a route to bifunctional derivatives or ethoxylates as nonionic surfactants.
Oligomerization of olefins is a major route that is currently used to make higher olefins, especially xcex1-olefins. Other routes to xcex1-olefins in decreasing importance are: paraffin wax cracking, paraffin dehydrogenation, and alcohol dehydration. A variety of catalysts for olefin oligomerization based on both early and late transition metals have been reported. One method uses Al-alkyls and is based on improvements to the original 1952 Ziegler""s Alfen process. Another example is the Shell Higher Olefin Process (SHOP), in which a nickel hydride catalyst is used. While these catalysts are highly active and effective, they normally only work with clean feeds. In other words, possible contaminants such as H2, CO, and especially mercaptans and thiophene, which are likely to exist in the feed, will poison the catalysts. Late transition metal catalysts are generally thought to be relatively insensitive to non.-sulfur-containing contaminants such as H2 and CO; however, sulfur-containing contaminants are potent poisons to late transition metal catalysts. Removing the contaminants and cleaning up the feed adds cost to current olefin oligomerization processes. It is therefore desirable to develop a catalyst that can work directly with crude feed, thereby offering potential cost-saving advantages to the process of olefin oligomerization.
Research in the area of olefin oligomerization has been very extensive, and there is continuing interest in developing new and more robust catalysts. Catalysts ranging from transition metal complexes, organoaluminum compounds, and Lewis and Brnsted type compounds to inorganic salts and oxides have been reported.
Compared to the large body of literature available on olefin oligomerization catalysts with nitrogen and/or phosphorous ligands, there have been very few reports in the open literature describing olefin oligomerization catalysts with sulfur ligands. Masters, et al. (J. Chem. Soc. Dalton Trans. 1993, 59-68; U.S. Pat. No. 4,533,651; WO 83/02907) reported a system that comprises a nickel(II) complex and a co-catalyst. The nickel(II) complex is a square-planar species with a phosphine or phosphite ligand, a halogen ligand, and a substituted dithio-xcex2-diketone ligand. The co-catalyst is an aluminum alkyl or an alkyl-aluminum chloride. The system catalyzes the oligomerization and/or isomerization of olefins. However, whether the system was contaminant-tolerant was not raised or discussed.
Japanese Patent No. 70007522 discloses a catalyst for co-oligomerization of butadiene and ethylene, wherein the catalyst consists of: (1) a metal complex having the formula M(S2C2Ph2)2, where M is Ni or Co and Ph is a phenyl radical; and (2) an organic Al compound of general formula R3Al or R2AlX, wherein R is a hydrocarbon radical and X is a halogen atom. A mixture of isomers, including C6-C8 dienes, C10 trienes, and vinylcyclohexane, is formed as a result of coupling between butadiene and ethylene, and the product ratio is dependent on the metal used. No oligomerization product from ethylene was observed even though ethylene of pressures up to 700 psig was used.
Riccardo Po, et al., Journal of Polymer Science: Part A: Polymer Chemistry, Vol. 36, 2119-2126 (1998), discloses the polymerization of styrene using catalytic systems based on nickel derivatives and methylaluminoxane. Even though compound Ni(Ph2C2S2)2 (labeled PS8b) was listed among the catalysts tested, it did not produce any polymeric product, indicating that the compound was inactive in polymerizing styrene.
Organometallic catalyst technology is also a viable tool in oligomerization processes that produce linear xcex1-olefins for use as feedstock in various other processes. However, one problem often encountered when using many of these catalyst systems is the propensity to produce xcex1-olefins with very low selectivity (i.e., a Schulz-Flory type distribution with high k values). For instance, many of the linear xcex1-olefins made today utilize a neutral nickel (II) catalyst having a planar geometry and containing bidentate monoanionic ligands. While these planar nickel (II) catalysts do produce linear xcex1-olefins, these catalysis systems exhibit a Schulz-Flory type of distribution over a very wide range (i.e., C4-C30+).
To address the Schulz-Flory distribution problem, chromium metal-based catalysts have become popular for use in certain oligomerization processes. More precisely, chromium complexes have been used to oligomerize ethylene in order to form linear xcex1-olefins with improved distributions. In fact, there have been reports of specific chromium catalysts that selectively trimerize ethylene to 1-hexene. See U.S. Pat. Nos. 5,814,575, 5,763,723 and 5,550,305. These techniques employ the use of a chromium compound in conjunction with aluminoxane along with one of a variety of compounds such as nitrites, amines and ethers. Unfortunately, while these techniques have been able to selectively produce xcex1-olefins, polymer is formed as a co-product. Of course, when polymer is co-produced, the yield of desirable xcex1-olefin products decreases accordingly. Also, as a practical matter, polymer build-up in the reaction vessel can severely hamper production efficiency, thereby limiting the commercial use of such processes.
As discussed above, the organometallic catalyst technology now being used to produce xcex1-olefins has two major disadvantages. First, many of the organometallic catalysts produce xcex1-olefins with a Schulz-Flory type distribution. Unfortunately this Schulz-Flory type distribution is not ideal when short chain xcex1-olefins are desired; in other words, the selectivity is not good enough to maintain efficient processes. Because xcex1-olefins are used as intermediates for specific products, xcex1-olefins with certain chain lengths are desired. For instance, the following are examples of xcex1-olefin chain lengths that would be desirable as feeds for certain product types: C4 to C8 for co-monomer in ethylene polymerization; C10 for lube quality poly-xcex1-olefins; and C12 to C20 for surfactant products. Thus, considerable inefficiency and waste are present when significant amounts of xcex1-olefins are produced having chain lengths outside of the range required for production of a particular chemical. Second, while some of the current organo-metallic catalysts may improve selectivity, most also produce polymer co-products. This lowers the yield of the desired product. Additionally, the polymer co-products can also accumulate in the reaction vessel. These disadvantages make commercial use of organo-metallic catalysts less attractive and inefficient. Hence, there is still a need for improving the selectively and efficiency of linear xcex1-olefin production.
The instant invention provides a method for selectively and efficiently producing higher olefins. The method includes contacting a feed comprising olefinic monomers under oligomerization conditions with a catalyst composition comprising a compound having the formula M[S2C2(RaRb)]2 or M[S2C6(R1R2R3R4)]2, wherein M is a late transition metal, Ra, Rb, R1, R2, R3 and R4 are independently selected and may be the same or different and are selected from hydrogen, electron withdrawing groups and unsubstituted and substituted hydrocarbyl groups; and an activating cocatalyst.
These and other features, aspects and advantages of the present invention will become better understood with regard to the following description and appended claims.
Generally speaking, the method of the invention selectively produces oligomers. The term xe2x80x9coligomersxe2x80x9d as used in this specification should be appreciated by one skilled in the art as meaning xcex1-olefins having approximately three to forty carbon atoms. More preferred xe2x80x9cshort chainxe2x80x9d linear xcex1-olefins have approximately three to twenty carbon atoms. Most preferred short chain linear xcex1-olefins have approximately four to twelve carbon atoms. The monomers used to produce the xcex1-olefins are olefinic monomers such as ethylene, propylene and butenes. Higher olefinic monomers, such as hexenes and octenes, may also be employed to produce the xcex1-olefins.
The invention provides a novel oligomerization method which utilizes a catalyst composition comprising a compound having the formula M[S2C2(RaRb)]2 or M[S2C6(R1R2R3R4)]2, wherein M is a late transition metal, Ra, Rb, R1, R2, R3 and R4 are independently selected and may be the same or different and are selected from hydrogen, electron withdrawing groups and unsubstituted and substituted hydrocarbyl groups; and an activating cocatalyst. Such compounds having formulas (I) M[S2C2(RaRb)]2 and (II) M[S2C6(R1R2R3R4)]2 are known in the art as metal dithiolene compounds. Dithiolene is a commonly used name for 1,2-enedithiolate or benzene-1,2-dithiolate and related dithiolates.
Compounds of formula (I) and (II) may be obtained by reacting a suitable source of metal with a suitable source of ligand, which may be obtained from commercial sources or prepared as described according to methods known in the art. The source of metal can also be obtained commercially. Typical examples of how to make the dithiolene compounds of formula (I) can be found in the literature, as disclosed in Schrauzer, et al., J.A.C.S., Vol. 87 (7), 1483-9 (1965), incorporated by reference herein. Anions of compounds of formula (II) can be made by reacting a metal salt with arene-1,2-dithiols and their analogs, as taught in Gray, et al., J.A.C.S., Vol. 88 (21), 4870-5 (1966), incorporated by reference herein. The neutral form of formula (II) is accessible by oxidation, either chemically or electrochemically. Additionally, compounds of formula (I) may be produced by the processes taught in U.S. Pat. No. 6,120,692 to Wang, et al., and Davison, et al., Inorganic Syntheses, Vol. 10, 8-26 (1967), both incorporated by reference herein. 
The structural formula for the transition metal bis(dithiolene) compounds are shown above in Formula (I) for M[S2C2(RaRb)]2 and in Formula (II) for M[S2C6(R1R2R3R4)]2.
In formula (I), M is a late transition metal, preferably either Fe, Co, Ni, Pd or Pt. Most preferably, M is Ni. Ra and Rb may be the same or different, and are independently selected from hydrogen, electron-withdrawing groups including those that are or contain heterocyclic, cyano, carboxylate, carboxylic ester, keto, nitro, and sulfonyl groups, and hydrocarbyl groups, including alkyl, cyclo alkyl, alkenyl and aryl groups, unsubstituted or fully or partly substituted. Preferably, Ra and Rb are substituted or unsubstituted hydrocarbyl groups. Most preferably, Ra and Rb are CF3, CH3 or phenyl groups.
As is known in the art, another type of dithiolene compound that may also be used contains benzene dithiolato ligands, represented by the structure in the formula (II) above. In the formula (II), M also is a late transition metal with the same characteristics as described for formula (I). R1, R2, R3, and R4 are independently selected and may be the same or different, and R1 to R4 are hydrogen, electron-withdrawing groups as described above, and unsubstituted or fully or partly substituted hydrocarbyl groups including alkyl, cycloalkyl, alkenyl, and aryl groups, preferably with substituents at the carbon atoms of the hydrocarbyl group that are electron-withdrawing groups. Preferably, the group is a halo group.
Those skilled in the art would recognize more complex forms in the dithiolene class that also may be used, as disclosed in Mueller-Westerhoff, U. T., xe2x80x9cDithiolene and Related Speciesxe2x80x9d, Comprehensive Coordination Chemistry, Vol. 2, 595-631 (1987), and McCleverty, J. A., xe2x80x9cMetal 1,2-Dithiolene and Related Complexesxe2x80x9d, Prog. Inorg. Chem., Vol. 10, 49-221 (1968), both incorporated by reference herein.
Thus, the compound can be any late transition metal bis(1,2-enedithiolate). The compound is preferably a substituted 1,2-enedithiolate where the substituent is an aryl or alkyl group, more preferably the alkyl group is a halo-alkyl group, and most preferably the compound is the bis(dithiobenzil) metal compound or the bis[1,2-bis(trifluoromethyl)ethylene-1,2-dithiolato] metal compound.
In one embodiment, the activating cocatalyst is selected from the group consisting of alkylaluminoxanes, aluminum alkyls, aluminum halides, alkyl aluminum halides, Lewis acids, alkylating agents, and mixtures thereof. In a more preferred embodiment, the activating cocatalyst is methylaluminoxane. The preferred ratio of metal compound to activating cocatalyst is from 1:10xe2x88x922 to 1:106.
Lewis acids other than any of the foregoing list and the mixtures of the foregoing can also be used in conjunction with alkylating agents, such as methyl magnesium chloride and methyl lithium. Examples of such Lewis acids are those compounds corresponding to the formula: Rxe2x80x3xe2x80x33B, where Rxe2x80x3xe2x80x3 independently each occurrence is selected from hydrogen, silyl, hydrocarbyl, halohydrocarbyl, alkoxide, aryloxide, amide or combination thereof, said Rxe2x80x3xe2x80x3 having up to 30 non-hydrogen atoms.
It is to be appreciated by those skilled in the art, that the above formula for the preferred Lewis acids represents an empirical formula, and that many Lewis acids exist as dimers or higher oligomers in solution or in the solid state. Other Lewis acids which are useful in the catalyst compositions of this invention will be apparent to those skilled in the art.
Other examples of suitable cocatalysts are discussed in U.S. Pat. Nos. 6,037,297 and 5,198,401, and PCT patent documents PCT/US97/10418 and PCT/US96/09764, all incorporated by reference herein.
The composition described above may also be supported. The support material is preferably a porous material which includes, but is not limited to, inorganic oxides, talc, and inorganic chlorides. The support material may also be resinous materials such as polystyrene polyolefin or polymeric compounds. The support material may also be any other organic material that has an average particle size greater than approximately 10 micron. These catalysts are generally physisorbed on the support. The catalysts can also be supported on mesoporous materials. In a more preferred embodiment, the composition is supported by silica.
The metal dithiolene compound of the invention is used in conjunction with the cocatalyst to oligomerize olefinic monomers. Thus, the invention also provides a method for producing linear xcex1-olefins by contacting olefinic monomers with the composition described above under certain temperature and pressure conditions conducive to forming oligomers, while minimizing, or totally eliminating, any polymer co-product. Olefinic monomers used for producing the linear xcex1-olefins include, but are not limited to, ethylene, propylene, butenes, and mixtures thereof. A preferred olefinic monomer is ethylene. In one embodiment, the invention produces linear xcex1-olefins having approximately four to twenty carbon atoms. In a more preferred embodiment, the invention produces xcex1-olefins having four to twelve carbon atoms.
The novel method advantageously oligomerizes olefinic monomers derived from crude feed streams that contain contaminants. Such contaminants include H2, CO, paraffins, C2H2, H2O, CO2 and alkynes. However, the most harmful, and least addressed in the art, contaminants are sulfur-containing compounds. The instant method successfully oligomerizes olefinic monomers into higher olefins even with crude feeds contaminated with sulfur-containing compounds, such as H2S, mercaptans, sulfides, thiophenes and derivatives thereof.
Generally, oligomerization may be accomplished utilizing similar temperatures and pressures used in the art. More specifically, temperature ranges from about xe2x88x92100 to 250xc2x0 C. and pressures from about 5 to 30,000 psig are acceptable. The most preferred temperature range is from about 0xc2x0 C. to 100xc2x0 C., while the most preferred pressure range is from about 15 to 2,000 psig.
Furthermore, oligomerization may take place in a solvent, neat (e.g., no solvent and liquid condensed olefin), or in a gas phase (e.g., olefin in gas phase and catalyst in the solid phase). When oligomerization is conducted in a solvent phase, suitable solvents include, but are not limited to, ethane, propane, butane, pentane, hexane, toluene, methylene chloride, carbon dioxide and mixtures thereof.
As for oligomerization in a gas phase, ExxonMobil Chemical Corporation""s gas phase catalyst technology, as described in U.S. Pat. No. 5,554,704, which is herein incorporated by reference, teaches a process for producing a supported catalyst. The supported catalyst can then be used in a solvent-free system wherein the gas phase xcex1-olefin is passed through a fixed bed of catalyst. The condensed xcex1-olefin product is then separated from the system.
The invention is further described in the following non-limiting examples.